Download Retinoid-induced apoptosis in normal and neoplastic tissues

Cell Death and Differentiation (1998) 5, 11 ± 19
 1998 Stockton Press All rights reserved 13509047/98 $12.00
Review
Retinoid-induced apoptosis in normal and neoplastic
tissues
Laszlo Nagy1,3,4, Vilmos A. Thomazy1, Richard A. Heyman2
and Peter J.A. Davies1,3
1
2
3
4
Department of Pharmacology, University of Texas-Houston, Medical School,
Houston, Texas 77225 USA
Ligand Pharmaceuticals, San Diego, California, 92121 USA
Corresponding author: PJAD, tel: 713-500-7480; fax: 713-500-7455;
e-mail: [email protected]
Present address for correspondence: The Salk Institute for Biological Studies,
Gene Expression Laboratory, La Jolla, California 92037;
tel: (619) 453-4100 fax:(619) 455-1349; e-mail: [email protected]
Received 18.8.97; revised 19.9.97; accepted 22.9.97
Edited by M. Piacentini
Abstract
Vitamin A and its derivatives (collectively referred to as
retinoids) are required for many fundamental life processes,
including vision, reproduction, metabolism, cellular differentiation, hematopoesis, bone development, and pattern
formation during embryogenesis. There is also considerable
evidence to suggest that natural and synthetic retinoids have
therapeutical effects due to their antiproliferative and
apoptosis-inducing effectsin human diseases such ascancer.
Therefore it is not surprising that a significant amount of
research was dedicated to probe the molecular and cellular
mechanisms of retinoid action during the past decade. One of
the cellular mechanisms retinoids have been implicated in is
the initiation and modulation of apoptosis in normal
development and disease. This review provides a brief
overview of the molecular basis of retinoid signaling, and
focuses on the retinoid-regulation of apoptotic cell death and
gene expression during normal development and in pathological conditions in vivo and in various tumor cell lines in vitro.
Keywords: retinoic acid; retinoic acid receptor; RAR; RXR;
apoptosis; tissue transglutaminase; gene expression
Abbreviations: ATRA, All-trans retinoic acid; 9-cis RA, 9-cis
retinoic acid; PG-J2, 15-deoxy-D12,14 PGJ2; RAR, retinoic acid
receptor; RXR, retinoid X receptor; Tgase, tissue transglutaminase; SMRT, silencing mediator of retinoid and thyroid
receptors; N-CoR, nuclear receptor corepressor; CREB, creb
binding protein; HDAC-1, histone deacetylase-1
Retinoids are modulators of transcription
The retinoid's mechanism of action in both normal and
pathological conditions has been the subject of more than a
decade of intense research. The cloning and discovery of the
retinoic acid receptor (RAR), which belongs to the superfamily
of ligand-activated transcription factors (nuclear receptors)
revolutionized our understanding as to how retinoids exert
their pleiotropic effects (for reviews see Chambon (1996);
Mangelsdorf et al (1994)). Members of the nuclear receptor
superfamily mediate the biological effects of many hormones,
vitamins and drugs (i.e. steroid hormones, thyroid hormones,
vitamin D, prostaglandin-J2 (PG-J2) and drugs that activate
peroxisomal proliferation). There are two families of retinoid
receptors, Retinoid X Receptors (RXRs) that bind 9-cis
retinoic acid (9-cis RA) and Retinoic Acid Receptors (RARs)
that bind both 9-cis RA and all-trans retinoic acid (ATRA) (for
reviews see Chambon 1996; Mangelsdorf et al, 1994)). Each
of these receptor families includes at least three distinct
genes, (RARa,b and g; RXRa,b and g) that through differential
promoter usage and alternative splicing, give rise to a large
number of distinct retinoid receptor proteins (for reviews see
Chambon 1996; Mangelsdorf et al, 1994).
There is a clear functional distinction between the RXRs
and the RARs. Nuclear receptors generally function as
dimers; steroid receptors form homodimers while most of the
other members of the superfamily form heterodimers with
other nuclear receptors. RXRs form heterodimers with
several nuclear receptors including thyroid hormone
receptors (TR), the vitamin D receptor (VDR) and the
receptors for the peroxisomal proliferator drugs and PG-J2
(PPARs). RXRs also heterodimerize with the RARs and can
homodimerize as well (Chambon, 1996; Mangelsdorf et al,
1994). Because of the multiplicity of RXR partners, ligands
that activate RXRs can have biological activity via diverse
endocrine signaling pathways. The function of RARs is more
restricted. RARs do not form homodimers but they do bind to
RXRs. Generally ligand activation of both the RAR and RXR
component of RAR/RXR heterodimers can contribute to the
activity of the receptors (Roy et al, 1995) although there are
some situations where ligand activation of either component
alone is sufficient for a full biological response (Kurokawa et
al, 1994; Nagy et al, 1995).
The role of ligands in the regulation of retinoid receptor
function is complex. Unliganded receptors bind to the
retinoid-response elements of retinoid-regulated genes, and
recruit transcription factors that negatively regulate gene
expression (Chen and Evans, 1995; Horlein et al, 1995).
Thus unliganded receptors can act as negative transcription
factors. Retinoids bind to a ligand-binding site in the
carboxyl-terminal half of the RARs (E/F domain) causing
marked conformational changes (Bourguet et al, 1995;
Renaud et al, 1995) by releasing the negative regulatory
factors (corepressors) and/or by facilitating the recruitment
of positive regulators (coactivators) of gene expression
(Kurokawa et al, 1995) (Figure 1). This way ligands can
directly activate the expression of target genes by either
relieving negative control or facilitating the activity of
Retinoid induced apoptosis
L Nagy et al
12
positive transcription factors. In addition to their direct
effects on transcription, ligand-activated RARs modulate
the activity of other transcription factors such as AP-1 (this
is termed cross-coupling) (Schule et al, 1991). Activated
RARs inhibit the activity of the transcription factors and
thereby control the expression of AP-1 regulated genes.
Thus there are two ways that retinoids can regulate gene
expression, either by direct effects on transcription or in an
indirect effect on AP-1 activity. The first mechanism will be
dependent on RRE' s within the target genes whereas the
latter mechanism will not. A surprising development in
retinoid biology has been the recent discovery that the two
mechanisms for the control of gene expression by retinoids
are the basis of different biological responses. The antiproliferative effects of retinoids are linked to the inhibition of
AP-1 activity (Chen et al, 1995; Fanjul et al, 1994; Nagpal
et al, 1995) whereas the induction of cellular differentiation
depends on the direct activation of transcription of specific
retinoid-regulated genes. Synthetic retinoids, such as
SR11220 (Fanjul et al, 1994), have been developed that
selectively activate the AP-1 inhibitory domains of retinoid
receptors without activating transcription. These compounds provide powerful pharmacologic tools for identifying AP-1 dependent aspects of retinoid activity in vivo.
The recent discovery of nuclear receptor associated
proteins (cofactors = coactivators and corepressors) provides us with detailes as to how DNA bound unliganded
and liganded receptor dimers influence transcription of
target genes both by direct and indirect mechanisms. For
the direct transcriptional effects a simple binary paradigm is
emerging from these studies (Figure 1). Unliganded
receptors bind to response elements of target genes and
repress transcription through recruitment of a repressor
complex containing corepressors (SMRT/N-CoR), Sin3 and
histone deacetylases (HDAC-1 and 2) (Alland et al, 1997;
Heinzel et al, 1997; Nagy et al, 1997b). This probably leads
to histone deacetylation and formation of an inactive
chromatin structure preventing transcription. Ligand binding causes the dissociation of corepressor proteins and
promotes association of coactivators with liganded receptors (Figure 1). Interestingly, several of the so far identified
coactivator proteins have histone acetylase activity i.e.
CBP/p300 (Ogryzko et al, 1996) and ACTR (Chen et al,
1997) which contributes to the formation of an active
chromatin structure and results in the transcription of the
target gene. Remarkably, several of the coactivators and
corepressors are shared by multiple signaling pathways
(i.e. CBP has been implicated in AP-1, pS3, STAT signaling
among others and Sin3, HDAC-1 are involved in Mad-Max
signaling (Ayer et al, 1995; Hassig et al, 1997; Kamei et al,
1996). This raises the possibility that formation of regulatory
(activator or repressor) complexes and/or competition for
limiting amounts of these proteins may prove to be critical
in determining which signaling pathway can be activated in
a given cell at a given time. This model of transcriptional
repression and activation by nuclear receptors and their
cofactors provides a direct link not only between multiple
signaling pathways critical in cellular proliferation, differentiation and apoptosis but also between these pathways
and the chromatin structure of target genes. It is likely
though that there are additional mechanisms (i.e. phosphorylation, direct interactions between receptors/cofactors
and the basal transcriptional machinery) associated with
cofactor/receptor function which may contribute to the fine
tuning of transcriptional regulation by nuclear receptors.
Role of retinoid receptors in vivo
Despite the strong evidence for the specialized functions of
RARs drawn from transfection and cultured cell studies,
experiments carried out in whole animals have suggested that
the relationship between the functions of the individual RARs
may be much more complex (for a review see Kastner et al
(1995)). Deletion of individual RARs in mice by homologous
recombination (receptor knock-outs) have produced surprisingly modest phenotypic consequences. The RARb-null
genotype is without an apparent phenotype (Luo et al,
1995). The knock-out of RARa1, the predominant RARa
isoform in the embryo, has no discernable phenotype (Li et al,
1993; Lufkin et al, 1993). Deletion of both RARa isoforms, the
RARa-null genotype, showed normal embryonic viability but
early post-natal death. The RARa-null animals are sterile due
Figure 1 Hormonal targeting of nuclear complexes to chromatin template. In the absence of hormone, a SMRT, Sin3A and HDAC1 complex associates with
unliganded receptor heterodimers. In this complex histone deacetylase activity creates a repressive chromatin environment. Addition of hormone triggers the
release of the repressor complex and recruitment of co-activators that include histone acetylases such as CBP/p300 and P/CAF. This results in local acetylation of
histones the recruitment of the basal transcription machinery and transcriptional activation. (Adapted from Perlmann and Evans, (1997))
Retinoid induced apoptosis
L Nagy et al
13
to testis degeneration and they also showed partial syndactyly
(evidence for deficient cell death in the interdigital webs of the
developing limbs) (Lufkin et al, 1993). Deletion of the RARg2
isotype is without evident phenotype whereas knock-out of
both RARg isoforms, the RARg-null phenotype, has normal
embryonic viability but early post-natal mortality due to growth
retardation. These animals are sterile due to prostatic
abnormalities, show subtle changes in vertebral patterning
and also show partial syndactyly (Kastner et al, 1995; Lohnes
et al, 1993). In the context of intact physiological systems,
deficits in the function of individual RARs appear to be
compensated for by the activity of the residual RARs present
in the affected tissues. This is likely to be the case since
double knock-outs, particularly of RARs a and g, produce a
much more severe phenotype than the single receptor knockouts (Lohnes, 1994; Mendelsohn, 1994). The phenotype of
the RARa, g-double null animals replicates most of the
teratogenic effects of vitamin A deficiency (VAD) (Wilson et
al, 1953; Wolbach and Howe, 1925) suggesting that the
combined activity of these two receptors is central to the
morphogenic effects of retinoids.
The results obtained with retinoid receptor knock-out
animals suggest that there may be redundancies in the
function of RARs in vivo. This redundancy may reflect the
fact that in vivo, different receptors may contribute to the
regulation of the same genes, i.e. the level of transcription
of those genes reflects the total level of receptors and
ligands present in the tissue rather than the presence of
any particular receptor (Kastner et al, 1995). This model of
redundancy is supported by the observation that defects in
retinoid-regulated gene expression in F-9 cells rendered
null for RARa can be partially reversed by over-expression
of RARg (Taneja et al, 1995). Similar results have been
obtained in studies of myeloid (HL-60) differentiation
(Robertson et al, 1992a,b). A retinoid-resistant cell line
(HL-60R) was developed by Collins and associates. This
cell line has a mutant allele of RARa (RARa411) which is a
truncated, dominant negative transcription factor inhibiting
retinoid signaling. Overexpression of either RARs or RXRs
restores retinoid responsiveness in these cells (Robertson
et al, 1992b).
An alternative explanation for the apparent redundancy
in RAR function in vivo is that it may represent overlapping
biological responses. Each RAR may preferentially regulate
specific retinoid-regulated genes that co-operate in producing a full biological response. The function of the individual
receptor may be sufficient to partially complete the process
but it is not as effective as the activity of both receptors
working in tandem. In the context of apoptosis occurring
during limb development, it is possible that the effects of
RARb may be directly on the apoptotic cells in the
interdigital webs (where it is expressed = cell autonomous)
whereas the effects of RARg, which is expressed in
adjacent connective tissues, may involve the induction of
cytokines that trigger the death of the cells in the interdigital
webs (non cell-autonomous). Under normal circumstances
both mechanisms may contribute to the death of the
interdigital tissues but in the experimental situation deletion
of one of the pathways results in a partial block in
apoptosis. This model is supported by the observation
that both the RARa- and RARg-null phenotypes include
partial syndactyly (Kastner et al, 1992).
Retinoid receptors and apoptosis
Retinoids have long been recognized to have major effects on
cellular proliferation and differentiation. Retinoids have a
broad spectrum of anti-proliferative activity in cultured cell
systems, particularly transformed cells and are used in the
therapy of certain hyper-proliferative, pre-malignant and
malignant diseases (Gudas et al, 1994). In addition retinoids
alter the differentiation of many cell types, inhibiting
squamous differentiation in epithelial cells (Jetten et al,
1990) and inducing the differentiation of myeloid cells
(Robertson et al, 1992b). These activities underlie their use
in the treatment of leukemias and certain squamous cancers.
Recently it has become recognized that retinoids also
regulate the expression of programmed cell death, inducing
the death of certain cell types (Martin et al, 1990) and
inhibiting apoptosis in others (Yang et al, 1993).
The link between retinoids and cell death was an
outgrowth of studies on the cellular basis of retinoid
teratogenesis. Retinoids are potent teratogens producing
a complex array of malformations in skeletal and neural
crest-derived structures (Kochar, 1967). In a series of
studies carried out by Sulik and co-workers it was
demonstrated that one of the reasons for retinoid-induced
teratogenesis was the expansion of zones of physiological
cell death (Sulik et al, 1988). This was particularly true in
the limb, where expanded zones of apoptosis in the
interdigital tissues and excessive death in the patternforming ectoderm (apical ectodermal ridge) lead to complex
limb malformations (Alles and Sulik, 1989; Sulik and
Dehart, 1988). Similar effects of retinoids have been
reported in !ower vertebrates (Ferretti and Geraudie,
1995). Administration of retinoids to limb apoptosisdefective mice (hammertoe strain) lead to partial restoration of interdigital apoptosis (Zakeri and Ahuja, 1994) and
exogenous retinoids applied to limb explants or cultured
interdigital cells cause extensive apoptosis (Jiang and
Kochhar, 1992; Lee et al, 1994). Similar mechanisms
contribute to the cranio-facial teratogenesis of retinoids
(Sulik et al, 1988) and their effects on development of the
CNS (Alles and Sulik, 1990; Sulik et al, 1988).
In addition to their effects on embryonic tissues retinoids
have also been implicated in the induction of cell death in
many tumor-derived cultured cell systems. Retinoids-induce
apoptosis in myeloid leukemia cells (Martin et al, 1990),
neuroblastoma cells (Melino et al, 1994), breast (Seewald
et al, 1995), ovarian (Krupitza et al, 1995) and cervical
cancer cells (Oridate et al, 1995) and many other types of
cells (Atencia et al, 1994; Corbeil et al, 1994; Kalemkerian
et al, 1995; Nakamura et al, 1995). In many of these
systems ligands that activate particular subsets of RARs
have proven to be particularly efficacious. Apoptosis of
tracheal epithelial cells is preferentially induced by RARa
agonists (Zhang et al, 1995), apoptosis of thymic
lymphocytes responds best to RARg ligands (Szondy et
al, 1997), co-stimulation of RARa and RARg induces
apoptosis in neuroblastoma cells (Melino et al, 1997), and
Retinoid induced apoptosis
L Nagy et al
14
ligand activation of RARb may play a particularly important
role in limb teratogenesis (Soprano et al, 1994). We have
recently shown that RXR ligands can also be very effective
inducers of apoptosis in myeloid leukemia cells (Nagy et al,
1996b; Nagy et al, 1995). Our understanding of retinoidinduced apoptosis in tumor cells has been complicated by
the recent observation that some of the most active
inducers of tumor cell apoptosis, such as 4-hydroxyphenylretinamide are not active in RAR-signaling pathways
suggesting that alternative pathways of retinoid signaling
may be critical for some forms of apoptosis (Sheikh et al,
1995). This issue is still unresolved though because there
are other reports claiming that this compound acts through
RARg (Fanjul et al, 1996).
At present it is not known to what extent these
observations represent the heterogeneity of factors that
can induce apoptosis in cultured cells, particularly tumor
cells, as opposed to reflecting the true complexity of
retinoid-regulated apoptosis as it occurs in vivo.
While there is substantial evidence that extrinsic retinoids
administered to animals can be highly teratogenic due at
least in part to the induction of cell death, it is much less clear
whether the retinoids normally present in tissues, endogenous retinoids, play a physiological role in regulating cell
death. The question can be best addressed by administering
retinoid antagonists to animals and observing effects on
normal embryonic development. Unfortunately until recently
potent and specific retinoid antagonists have not been
available so this experimental approach has not been
explored. The recent development of RAR-antagonists,
AGN 193109 a very potent pan-RAR antagonist (Johnson
et al, 1995) and Ro13-5320, an RARa selective antagonist
(Apfel et al, 1992), have made such experiments feasible.
Although RAR antagonists have not been available, two
other approaches, deletion of individual retinoid receptors
and the induction of vitamin A deficiency, have provided
critical information on the role of retinoids as physiological
regulators of apoptosis. The Chambon laboratory has carried
out comprehensive studies on morphogenesis and cell death
in animals in which individual retinoid receptors have been
deleted by homologous recombination (receptor knock-out
animals, reviewed in (Kastner et al, 1995). Single receptor
knock-outs have a limited phenotype. However partial
syndactyly (a failure of regression of interdigital tissues due
to defective apoptosis) was a common feature of animals null
for either RARa or RARg (Kastner et al, 1995) suggesting
that either or both receptors were essential for normal cell
death in this tissue. These studies demonstrate that retinoid
receptors are crucial for normal apoptosis in the limb but do
not answer the question of whether the ligands for these
receptors, i.e. endogenous retinoids, are also required.
The role of endogenous retinoids in normal morphogenesis has also been approached by making animals retinoid
deficient via dietary deprivation (the vitamin A deficiency
syndrome ± VAD). A series of studies in rodents carried out
in the late 1940s established that VAD is very teratogenic,
resulting in malformations in the eye, cardiovascular system
and uro-genital system. Limb development abnormalities
were not noted (Wilson et al, 1953). Unfortunately these
studies preceded the recognition of apoptosis as a
component of normal morphogenesis and so the findings
were discussed in terms of abnormal patterns of cellular
differentiation rather than the possible failures in normal
tissue regression. Since that time there is no evidence in
the literature to suggest that the association between VAD
and apoptosis has been systematically addressed.
The available evidence suggests that retinoids are
involved in the regulation of apoptosis but they shed little
information on how these effects occur.
Retinoid regulated gene expression during
apoptosis: regulation of tissue transglutaminase
Retinoids produce their biological effects by alterations in
patterns of gene expression. Retinoids have been shown to
regulate the expression of regulatory factors of apoptosis
such as p21 (Lin et al, 1996) (Boccia et al, 1997) and Bcl-2
(Nagy et al, 1996b) as well as effector enzymes such as
transglutaminases (Chiocca et al, 1988) and sphingomyelinases (Riboni et al, 1995) which have been implicated in the
induction and execution of cell death. The normal and external
retinoid- induced in vivo expression pattern of one of the
retinoic acid receptors (RARb) suggested that it may be
associated with cell death also (Mendelsohn et al, 1991).
Analysis of the expression pattern of RARb in embryos from
mothers treated with teratogenic doses of ATRA, indicated
that mRAR-beta 2 promoter is selectively induced. These
findings suggest that overexpression of the mRAR-beta 2
isoform is involved in RA-generated malformations. This
raises the possibility that mRAR-beta 2 may have a role in
development of the limbs, as an inhibitor of cartilage
formation, in programmed cell death and in the formation of
loose connective tissue.
One of the most extensively studied systems is the
differentiation and subsequent apoptosis of the myeloid
leukemia cell line (HL-60). The initial observation that
retinoids induce apoptosis subsequent of cellular differentiation was made by Martin et al (1990). All-trans retinoic
acid (ATRA) induces differentiation of HL-60 cells toward
mature neutrophil granulocytes which subsequently die by
apoptotic cell death (Figure 2). Using receptor selective
retinoids we were able to demonstrate that the biological
response induced by ATRA a pan-receptor agonist (under
tissue culture conditions) has two components. The first
one is an RARa-induced differentiation phase which is
followed by an RXR-ligand-dependent apoptotic phase. If
one assumes that the receptor species mediating the
effects of the panagonist and RXR-specific retinoids on
apoptosis is also an RAR/RXR heterodimer, then it is likely
that ligand activation of the RXR moiety of this complex
alters the transciption of genes critical to the induction of
apoptosis. Alternatively, ligand-activation of different RXR/
nuclear receptor heterodimers or RXR homodimers may
regulate the apoptotic response. There are several changes
in the levels of expression of numerous apoptosis related
genes during the retinoid-induced death of HL-60 cells. The
level of the anti-apoptotic protein, Bcl-2 is down regulated
(Nagy et al, 1996b) whilst the expression levels of effector
enzymes such as tissue transglutaminase (Nagy et al,
Retinoid induced apoptosis
L Nagy et al
15
Differentiation RAR Mediated
Apoptosis RXR Mediated
Figure 2 Retinoid regulation of differentiation and apoptosis in myeloid leukemia cells (HL-60). HL-60 cell undergo retinoic acid induced differentiation and
subsequently die by apoptotic cell death. Upper panel: typical morphology of cells undergoing granulocytic differentiation (an RAR mediated process): myeloblast,
metamyelocyte, band, and segmented polymorphonuclear cells (from left to right). Lower panel: morphological changes associated of HL-60 cells undergoing
apoptosis (an RXR mediated effect): differentiated cell, nuclear fragmentation, and nuclear and cytoplasmic fragmentation and condensation, apoptotic remnant
(from left to right)
A
B
Figure 3 Retinoid regulation of tissue transglutaminase gene expression, the role of a versatile, tripartite response element (mTGRRE1) (A) RAR-RXR
heterodimers are able to bind to B and C halfsites (DR5, direct repeat) of mTGRRE1, upon ligand activation of either the RAR side or both RAR and RXR sides
transcription is initiated. (B) the DR5 element is also able to bind RXR homodimers, and confer RXR-dependent transcriptional activation upon ligand stimulation.
Two additional elements contribute to retinoid regulation of the mouse tissue transglutaminase gene: the A halfsite, which forms a DR7 with B and a region highly
homologous in mouse and human HR-1 (homology region-1) (Nagy et al, 1996a) which is located further downstream from mTGRRE1 and necessary for full
retinoid response in the context of the mouse tissue transglutaminase promoter
Retinoid induced apoptosis
L Nagy et al
16
1996b), ICE (caspase 1) and CPP32 (caspase 3) (Watson
et al, 1997) are elevated. It is interesting to note that Bcl-2
down-regulation is RAR-mediated while induction of tissue
transglutaminase is predominantly RXR-induced in this
myeloid cell line (Nagy et al, 1996b). This suggests that
differentiation and apoptotic cell death of these cells are
intimately linked and that retinoids are able to initiate an
orchestrated process which involves both positive and
negative changes in gene expression of a diverse group of
factors known to be associated with apoptosis.
We studied the transcriptional regulation of the apoptosis
effector tissue transglutaminase in order to gain insight into
the molecular mechanisms of retinoid regulated expression
of this enzyme. We attempted to identify cis-acting
regulatory elements which are able to direct expression
upon many diverse regulatory stimuli.
Tissue transglutaminase (TGase) is an intracellular
protein cross-linking enzyme that accumulates to high
levels in cells undergoing apoptotic cell death (for reviews
see Aeschlimann and Paulsson (1994); Greenberg et al
(1991); Nagy et al (1994)). The enzyme is both induced
and activated (by falling GTP and rising intracellular Ca2+)
during apoptosis (Fesus et al, 1987, 1989). Activation of
TGase in apoptotic cells, results in the extensive crosslinking of intracellular proteins during the fragmentation
phase of the apoptotic program (Fesus et al, 1989). Overexpression of TGase in cells (by transfection with TGase
expression vectors) results in high levels of spontaneous
apoptosis (Gentile et al, 1992; Melino et al, 1994). Inhibition
of TGase expression (with antisense expression vectors)
results in a reciprocal decrease in apoptotic activity (Melino
et al, 1994). It is thought that the formation of a densely
cross-linked intracellular matrix contributes to the fragmentation of cells during apoptosis as well as preventing the
leakage of intracellular proteins during the formation of
apoptotic bodies (Piredda et al, 1997).
TGase is expressed in many sites of physiological
apoptosis in vivo. During limb development, the enzyme
accumulates in clusters of apoptotic cells in the interdigital
webs as well as in the hypertrophic chondrocytes of the
developing bones (Thomazy and Davies, unpublished
observations). TGase also accumulates in the apoptotic
hepatocytes formed during the reversal phase of druginduced hepatomegaly (Thomazy and Fesus, 1989) as well
as in involuting mammary and prostate glands following
removal of trophic stimuli (Fesus et al, 1991). Comparison
of the results obtained with in situ hybridization probes and
TGase immunochemistry suggest that TGase mRNA is
expressed in cells prior to the development of morphologic
evidence of apoptosis whereas the actually enzyme
accumulates to high levels in cells that are well advanced
in the apoptotic program (Thomazy and Davies, unpublished observations). Studies with the TGase transgene
confirm that the promoter for the TGase gene is activated in
cells before the appearance of an apoptotic phenotype
(Nagy et al, 1997a) suggesting that the activation of the
gene is an early component in the commitment of cells
undergoing apoptosis.
The induction of TGase expression during in vivo
apoptosis can be replicated in some, but not all forms, of
apoptosis induced in vitro. The induction of apoptosis in
hepatocytes (Piacentini et al, 1992), neuroblastoma cells
(Melino et al, 1994), myeloid leukemia cells (Nagy et al,
1996b), trachealepithelial cells (Zhang et al, 1995) and
others has been associated with the induction of tissue
TGase expression. The very rapid apoptosis that develops
following DNA damage, a p53-dependent form of apoptosis, is not associated with evidence of increased TGase
expression (Lu and Davies, unpublished observations). In
summary it appears that the enzyme is induced in most
forms of physiological cell death, particularly in cell death
associated with morphogenesis and endocrine-regulated
tissue remodelling.
The association of TGase with apoptosis has lead to
considerable interest in the factors that regulate the
enzyme's expression. Several years ago we and others
demonstrated that retinoids were potent and specific
regulators of TGase expression in vitro and in vivo
(Davies et al, 1985; Moore et al, 1984). The effects of
retinoids are physiologically significant since depletion of
endogenous retinoids in rats, by vitamin A deficiency,
results in a global decrease in the levels of TGase
expressed in cells and tissues (Verma et al, 1992). We
subsequently cloned the mouse and human TGases and
showed that retinoids regulate transcription of the TGase
gene (Chiocca et al, 1988). An unusual feature of the
transglutaminase gene that it is under complex retinoid
regulation. It can be activated by ligands of both RARs and
RXRs in different cells and tissues (Nagy et al, 1996b).
This on one hand provides a mechanism so that the
enzyme's expression can be induced multiple forms of
retinoid induced apoptosis on the other it raises the
possibility that there may be multiple, separate cis-acting
elements contributing to the gene's regulation. More
recently we have isolated and characterized the promoters
for both the human and mouse Tgases (Lu et al, 1995;
Nagy et al, 1997a). We have identified in the mouse gene a
specific tripartite retinoid response element (mTG RRE-1)
(Figure 3) whose presence is required for efficient retinoiddependent activation of the promoter (Nagy et al, 1996a).
The TGase RRE-1 functions as a ligand-dependent
enhancer element capable of activating homologous and
heterologous promoters in a position- and orientationindependent manner (Figure 3). The promoter can be
activated by all three sub-types of RARs and in myeloid
leukemia cells by ligands that activate endogenous RXRs
(Beard et al, 1994; Nagy et al, 1996b). This profile matches
that of the endogenous gene.
The isolation of the TGase promoter coupled with the
observation that the gene was induced in many apoptotic
cells gave us an opportunity to ask an intriguing question.
Is it possible to use the TGase promoter as a targeting
vector capable of selectively inducing gene expression in
apoptotic cells? We coupled a 3.8 kb fragment of the
mouse TGase promoter to a beta-galactosidase reporter
gene and then used this reporter transgene to make
transgenic mouse lineages (Nagy et al, 1997a). In the
lineages we characterized, the transgene is selectively
expressed in the embryonic limb in regions of cells
undergoing apoptosis (Nagy et al, 1997a). The reporter
Retinoid induced apoptosis
L Nagy et al
17
gene can be detected in morphologically normal cells
associated with apoptotic cells as well as in cells that are
distinctly apoptotic suggesting that the transgene is actually
activated early in the apoptotic program (Nagy et al,
1997a). These studies have confirmed our hypothesis that
the activity of the TGase promoter is specifically linked to
regions (i.e. interdigital web) of apoptotic cell death in vivo.
In addition they demonstrate that all of the information
required to express this gene in apoptotic cells is
embedded within the proximal 3.8 kb of the promoter. This
observation taken together with our previous finding, that
retnoids are key regulators of the transcriptional activity of
the TGase promoter provides the foundation for our
proposal that retinoid regulation of tissue transglutaminase
and apoptosis likely to be coupled in vivo.
Perspectives
During the past 10 years there has been significant advance
in our understanding of how retinoids work and exert their
biological effects. Several aspects of the molecular mechanisms of retinoid receptor signaling has been clarified: nuclear
receptors for retinoids were cloned, natural and synthetic
ligands were identified and recently several receptor
interacting cofactors were discovered. Connections between
retinoids and the regulation of complex processes such as
differentiation and apoptotic cell death has been made both in
normal development and in disease. Thousands of synthetic
receptor specific retinoid analogs have been developed and
tested to elucidate the function and biological activites of
retinoid receptors. This has led to the development of new
therapeautic approaches in the treatment of certain leukemias, squamous cell carcinomas, breast and lung cancers as
well as in a variety of skin diseases. Several of these
treatments utilizes the fact that retinoids are able to induce
terminal differentiation and/or apoptosis of unwanted pools of
malignant or rapidly proliferating cells. Although a connection
between retinoids and apoptosis has been established, most
of the reports on this subject concentrate on documenting the
biological processes rather than dissecting the molecular
components of them. More mechanistic studies are needed to
clarify the molecular details of retinoid regulated apoptosis.
For instance, we do not fully understand the role of
endogenous retinoids and individual receptors as well as
the contribution of other signaling pathways to morphogenic
apoptosis. Also, identification of new target genes and
dissection of the mechanisms of their retinoid regulation
would shed light on the secondary regulatory components
and effector elements of retinoid modulated apoptosis.
Finally, it is still not clear how retinoid regulated gene
expression leads to the activation of the well characterized
executionary phase of apoptosis.
References
Aeschlimann D and Paulsson M (1994) Transglutaminases: protein cross-linking
enzymes in tissues and body fluids. Thromb Haemost 71: 402 ± 415
Alland L, Muhle R, Hou H, Potes J, Chin L, Schreiber-Agus N, and DePinho R (1997)
Nature 387: 49 55
Alles AJ and Sulik KK (1989) Retinoic-acid induced limb-reduction defects:
perturbation of zones of programmed cell death as a pathogenetic
mechanism. Teratology 40: 163 ± 171
Alles AJ and Sulik KK (1990) Retinoic acid-induced spina bifida: evidence for a
pathogenetic mechanism. Development 108: 73 ± 81
Apfel C, Bauer F, Crettaz M, Forni L, Kamber M, Kaufman F, LeMotte P, Pirson W and
Klaus M (1992) A retinoic acid receptor a antagonist selectively counteracts
retinoic acid effects. Proc Natl Acad Sci USA 89: 7129 ± 7133
Atencia R, Garcia-Sanz M, Unda F and Arechaga J (1994) Apoptosis during retinoic
acid-induced differentiation of F9 embryonal carcinoma cells. J Exp Cell Res 214:
663 ± 637
Ayer D, Lawrence Q and Eisenman R (1995) Mad-Max transcriptional repression is
mediated by ternary complex formation with mammalian homologs of yeast
repressor Sin3. Cell 80: 767 ± 776
Beard RL, Gil DW, Marler DK, Henry E, Colon DF, Gillett SJ, Davies PJA and
Chandraratna RAS (1994) Structural basis for the differential RXR & RAR activity
of stilbene retinoid analogs. Bioorganic and Medicinal Chemistry Letters 4:
1447 ± 1452
Boccia M, Xu Q, Wesley U, Xu Y, Korontsvit T, Loganzo F, Albino A and Scheinberg D
(1997) Mudulation of p53, WAF1/p21 and BCL-2 expression during retinoic acidinduced differentiation of NB4 promyelocytic cells. Leuk Res 21: 439 ± 447
Bourguet W, Ruff M, Chambon P, Gronemeyer H and Moras D (1995) Crystal
structure of the ligand-binding domain of the human nuclear receptor RXR-alpha.
Nature 375: 377 ± 382
Chambon P (1996) A decade of molecular biology of retinoic acid receptors. FASEB
Journal 10: 940 ± 954
Chen H, Lin R, Schiltz R, Chakravarti D, Nash A, Nagy L, Privalsky M, Nakatani Y and
Evans R (1997) Nuclear receptor coactivator ACTR is a novel histone
acetyltransferase and forms a multimeric activation complex with P/CAF and
CBP/p300. Cell 90: 569 ± 580
Chen J, Penco S, Ostrowski J, Balaguer P, Pons M, Starrett J, Reczek P, Chambon P
and Gronemeyer H (1995) RAR-specific agonist/antagonists which dissociate
transactivation and AP1 transrepression inhibit anchorage-independent cell
proliferation. EMBO J 14: 1187 ± 1197.
Chen JD and Evans RM (1995) A Transcriptional Co-Repressor That Interacts With
Nuclear Hormone Receptors. Nature 377: 454 ± 457.
Chiocca EA, Davies PJ and Stein JP (1988) The molecular basis of retinoic acid
action. Transcriptional regulation of tissue transglutaminase gene expression in
macrophages. J Biol Chem 263: 11584 ± 11589
Corbeil J, Rapaport E, Richman D and Looney D (1994) Antiproliferative effect of
retinoid compounds on Kaposi's sarcoma cells. J Clin Invest 93: 1981 ± 1986
Davies PJ, Murtaugh MP, Moore WT, Jr, Johnson GS and Lucas D (1985) Retinoic
acid-induced expression of tissue transglutaminase in human promyelocytic
leukemia (HL-60) cells. J Biol Chem 260: 5166 ± 5174
Fanjul A, Dawson M, Hobbs P, Jong L, Cameron J, Harlev E, Graupner G, Lu X and
Pfahl M (1994) A new class of retinoids with selective inhibition of AP- 1 inhibits
proliferation. Nature 372: 107 ± 111
Fanjul A, Delia D, Pierotti M, Rideout D, Qin J and Pfahl M (1996) 4-Hydroxyphenyl
retinamide is a highly selective activator of retinoid receptors. J Biol Chem 271:
22441 ± 22446
Ferretti P and Geraudie J (1995) Retinoic acid-induced cell death in the wound
epidermis of regenerating zebrafish fins. Dev Dyn 202: 271 ± 283
Fesus L, Thomazy V and Falus A (1987) Induction and activation of tissue
transglutaminase during programmed cell death. FEBS Lett 224: 104 ± 108
Fesus L, Thomazy V, Autuori F, Ceru MP, Tarcsa E and Piacentini M (1989) Apoptotic
hepatocytes become insoluble in detergents and chaotropic agents as a result of
transglutaminase action. FEBS Lett 245: 150 ± 154
Fesus L, Davies PJA and Piacentini M (1991) Apoptosis: molecular mechanisms in
programmed cell death. European Journal of Cell Biology 56: 170 ± 177
Gentile V, Thomazy V, Piacentini M, Fesus L and Davies PJ (1992) Expression of
tissue transglutaminase in Balb-C 3T3 fibroblasts: effects on cellular morphology
and adhesion. J Cell Biol ll9: 463 ± 474
Greenberg CS, Birckbichler PJ and Rice R H (1991) Transglutaminases:
multifunctional cross-linking enzymes that stabilize tissues. FASEB J 5:
3071 ± 307
Gudas L, Sporn M and Roberts A (1994) Cellular biology and biochemistry of the
retinoids. In The Retinoids. Biology, Chemistry and Medicine, M Sporn, A
Roberts and D Goodman, eds (New York: Raven Press)
Retinoid induced apoptosis
L Nagy et al
18
Hassig C, Fleischer T, Billin A, Schreiber S and Ayer D (1997) Histone deacetylase
activity is required for full transcriptional repression by mSin3A. Cell 89: 341 ±
347
Heinzel T, Lavinsky R, Mullen T-M, Soderstrom M, Laherty C, Torchia J, Yang W-M,
Brard G, Ngo S, Davie J, Glass C and Rosenfeld M (1997) N-CoR, mSin3 and
histone deacetylase-containing complexes in nuclear receptor and Mad
repression. Nature 387: 43 ± 48
Horlein AJ, Naar AM, Heinzel T, Torchia J, Gloss B, Kurokawa R, Ryan A, Kamel Y,
Soderstrom M, Glass CK and Rosenfeld MG (1995) Ligand-Independent
Repression By the Thyroid Hormone Receptor Mediated By a Nuclear Receptor
Co-Repressor. Nature 377: 377 ± 404
Jetten A, Kim J, Sacks P, Rearick J, Lotan D, Hong W and Lotan R (1990) Inhibition of
growth and squamous-cell differentiation markers in cultured human head and
neck squamous carcinoma cells by beta-all-trans retinoic acid. Int J Cancer 45:
195 ± 202
Jiang H and Kochhar DM (1992) Induction of tissue transglutaminase and apoptosis
by retinoic acid in the limb bud. Teratology 46: 333 ± 40
Johnson A, Klein E, Gillett S, Wang L, Song T, Pino M and Chandraratna R (1995)
Synthesis and characterization of a highly potent and effective antagonist of
retinoic acid receptors. J Med Chem 38: 4764 ± 4767
Kalemkerian G, Slusher R, Ramalingam S, Gadgeel S and Mabry M (1995) Growth
inhibition and induction of apoptosis by fenretinide in small -cell lung cancers cell
lines. J NCI 87: 1676 ± 1680
Kamei Y, Xu L, Heinzel T, Torchia J, Kurokawa R, Gloss B, Lin S, Heyman R, Rose D,
Glass C and Rosenfeld M (1996) A CBP integrator complex mediates
transcriptional activation and AP-1 inhibition by nuclear receptors. Cell 85:
403 ± 414
Kastner P, Mark M and Chambon P (1995) Nonsteroid nuclear receptors ± What are
genetic studies telling us about their role in real lif. Cell 83: 859 ± 869
Kochar D (1967) Teratogenic activity of retinoic acid. Acta Pathologica and
Microbiologica Scandinavica 70: 398 ± 404
Krupitza G, Hulla W, Harant H, Dittrich E, Kallay E, Huber H, Grunt T and Dittrich C
(1995) Retinoic acid induced death of ovarian carcinoma cells correlates with cmyc stimulation. Int J of Cancer 61: 649 ± 657
Kurokawa R, DiRenzo J, Boehm M, Sugarman J, Gloss B, Rosenfeld MG, Heyman
RA and Glass CK (1994) Regulation of retinoid signalling by receptor polarity and
allosteric control of ligand binding. Nature 371: 528 ± 531
Kurokawa R, Soderstrom M, Horlein A, Halachmi S, Brown M, Rosenfeld MG and
Glass CK (1995) Polarity-Specific Activities Of Retinoic Acid Receptors
Determined By a Co-Repressor. Nature 377: 451 ± 454
Lee KKH, Li FCH, Yung WT, Kung JLS, Ng JL and Cheah KSE (1994) Influence Of
Digits, Ectoderm, and Retinoic Acid On Chondrogenesis By Mouse Interdigital
Mesoderm In Culture. Developmental Dynamics 201: 297 ± 309
Li E, Sucov H, Lee K, Evans R and Jaenisch R (1993) Normal development and
growth of mice carrying a targeted disruption of the alpha 1 retinoic acid receptor
gene. Proc Natl Acad Sci USA 90: 1590 ± 1594
Lin M, Iawarone A and Freedman L (1996) Retinoid induction of U937 cell
differentiation: transcriptional activation of the human p21WAFI/CIP1 gene by
retinoic acid receptor. J Biol Chem 271: 31723 ± 31728
Lohnes D, Kastner P, Dierich A, Mark M, LeMeur M and Chambon P (1993) Function
of retinoic acid receptor gamma in the mouse. Cell 73: 643 ± 658
Lu S, Saydak M, Gentile V, Stein JP and Davies PJA (1995) Isolation and
characterization of the human tissue transglutaminase gene promoter. J. Biol.
Chem., 271: 9748 ± 9756
Lufkin T, Lohnes D, Mark M, Dierich A, Gorry M, Gaub P, LeMeur M and Chambon P
(1993) High post natal lethality and testis degeneration in retinoic acid receptor
alpha mutant mice. Proc Natl Acad Sci 90: 7225 ± 7229
Luo J, Pasceri P, Conlon R, Rossant J and Giguere V (1995) Mice lacking all iso-forms
of retinoic receptor beta develop normally and are susceptible to the teratogenic
effects of retinoic acid. Mech Dev 53: 61 ± 71
Mangelsdorf DJ, Umesono K and Evans RM (1994) The retinoid receptors. In The
retinoids, MB Sporn, AB Roberts and DS Goodman, eds (New York: Raven
Press), pp. 319 ± 350
Martin SJ, Bradley JG and Cotter TG (1990) HL-60 cells induced to differentiate
towards neutrophils subsequently die via apoptosis. Clinical Experimental
Immunology 79: 448 ± 453
Melino G, Annicchiarico-Petruzzelli M, Piredda L, Candi E, Gentile V, Davies PJ and
Piacentini M (1994) Tissue transglutaminase and apoptosis: sense and
antisense transfection studies with human neuroblastoma cells. Mol Cell Biol
14: 6584 ± 6596
Melino G, Draoui M, Bellincampi L, Bernassola F, Bernardini S, Piacentini M, Reichert
U and Cohen P (1997) Retinoic acid receptors a and g mediate the induction of
tissue transglutaminase activity and apoptosis in human neuroblastoma cells.
Exp Cell Res 235: 55 ± 61
Mendelsohn C, Ruberte E, LeMeur M, Morriss-Kay G and Chambon P (1991)
Developmental analysis of the retinoic acid -inducible RAR-beta 2 promoter in
transgenic animals. Development 113: 723 ± 34
Moore WT, Jr, Murtaugh MP and Davies PJ (1984) Retinoic acid -induced expression
of tissue transglutaminase in mouse peritoneal macrophages. J Biol Chem 259:
12794 ± 802
Nagpal S, Athanikar J and Chandraratna R (1995) Separation of transactivation and
AP1 antagonism functions of retinoic acid receptor alpha. J Biol Chem 270: 923 ±
927
Nagy L, Thomazy VA and Davies PJA (1994) Tissue transglutaminase: an effector in
physiologic cell death. Cancer Bulletin 46: 136 ± 140
Nagy L, Thomazy VA, Shipley GL, Fesus L, Lamph W, Heyman RA, Chandraratna
RAC and Davies PJA (1995) Activation of Retinoid X Receptors (RXR) induces
apoptosis in HL-60 cell lines. Mol Cell Biol 15: 3540 ± 3551
Nagy L, Saydak M, Shipley N, Lu S, Basilion JP, Yan ZH, Syka P, Chandraratna R,
Heyman R and Davies PJA (1996a) Identification and characterization of a
versatile retinoid response element (RARE/RXRE) in the promoter of the mouse
tissue transglutaminase gene. J Biol Chem 271: 4355 ± 4365
Nagy L, Thomazy VA, Heyman RA, Chandraratna RAS and Davies PJA (1996b)
Retinoid-regulated expression of BCL-2 and tissue transglutaminase during
differentiation and apoptosis of human myeloid leukemia (HL-60) cells.
Leukemia Research 19: 499 ± 505
Nagy L, Thomazy V, Saydak M, Stein J and Davies P (1997a) The promoter of the
mouse tissue transglutaminase gene directs tissue-specific, retinoid-regulated
and apoptosis-linked expression. Cell Death Differ 4: 534 547
Nagy L, Kao H-Y, Chakravarti D, Lin R, Hassig C, Schreiber S, Ayer D and Evans R
(1997b) Nuclear receptor repression mediated by a complex containing SMRT,
mSin3A, and histone deacetylase. Cell 89: 373 ± 380
Nakamura N, Shidoji Y, Yamada Y, Hatakeyama H, Moriwaki H and Muto Y (1995)
Induction of apoptosis by a cyclic retinoid in the human hepatoma derived cell line
Huh-7. Biochem. Biophys. Res. Com. 207: 328 ± 388
Ogryzko V, Schiltz R, Russanova V, Howard B and Nakatani Y (1996) The
transcriptional coactivators p300 and CBP are histone acetyltransferases. Cell
87: 953 ± 959
Oridate N, Lotan D, Mitchell M, Hong W and Lotan R (1995) Induction of apoptosis by
retinoids in human cervical carcinoma cell lines. International J of Oncology 7:
433 ± 441
Perlmann T and Evans R (1997) Nuclear receptors in Sicily: all in the famiglia. Cell 90:
391 ± 397
Piacentini M, Ceru MP, Dini L, Di Rao M, Piredda L, Thomazy V, Davies PJA and
Fesus L (1992) In vivo and in vitro induction of tissue transglutaminase in rat
hepatocytes by retinoic acid. Biochimica et Biophysica Acta, 171 ± 179
Piredda L, Amendolo A, Colizzi V, Davies P, Farrace MG, Fraziano M, Gentile V, Uray
I, Piacentini M and Fesus L (1997) Lack of tissue transglutaminase protein crosslinking leads to leakage of macromolecules from dying cells: relationship to
development of autoimmunity in MRL Ipr/lpr mice. Cell Death Differ 4: 463 ± 472
Renaud J, Rochel N, Ruff M, Vivat V, Chambon P, Gronemeyer H and Moras D (1995)
Crystal structure of the RAR gamma ligand binding domain bound to all trans
retinoic acid. Nature 378: 681 ± 689
Riboni L, Prinetti A, Bassi R, Caminiti A and Tettamanti G (1995) A mediator role of
ceramide in the regulation of neuroblastoma Neuro2a cell differentiation. J Biol
Chem 270: 26868 ± 26875
Robertson KA, Emami B and Collins SJ (1992a) Retinoic acid-resistant HL-60R cells
harbor a point mutation in the retinoic acid receptor ligand-binding domain that
confers dominant negative activity. Blood 80: 1885 ± 1889
Robertson KA, Emami B, Mueller L and Collins SJ (1992b) Multiple members of the
retinoic acid receptor family are capable of mediating the granulocytic
differentiation of HL-60 cells. Mol. Cell. Biol. 12: 3743 ± 3749
Roy D, Taneja R and Chambon P (1995) Synergistic activation of retinoic acid
responsive genes and induction of embryonal carcinoma cell differentiation by an
RARalpha RARbeta or RARgamma-selective ligand in combination with retinoid
X receptor-selective ligand. Mol Cell Biol 15: 6481 ± 6487
Schule R, Rangarajan P, Yang N, Kliewer S, Ransone L, Bolado J, Verma I and Evans
R (1991) Retinoic acid is a negative regulator of AP-1-responsive genes. Proc
Natl Acad Sci USA 88: 6092 ± 6096
Retinoid induced apoptosis
L Nagy et al
19
Seewald V, BS Parker M, Collins S and Swisshelm K (1995) Expression of retinoic
acid beta mediates retinoic acid-induced groth arrest and apoptosis in breast
cancer cells. Cell Growth and Diff 6: 1077 ± 1088
Sheikh M, Shao Z, Li X, Ordonez J, Conley B, Wu S, Dawson M, Han Q, Chao W, Quick
T, Niles R and Fontana J (1995) N-(4-Hydroxyphenyl)Retinamide(4-Hpr-)Mediates biological actions involve retinoid receptor-independent pathways in
human breast carcinoma. Carcinogenesis 16: 2477 ± 2486
Soprano D, Gyda M, Jiang H, Harnish D, Ugen K, Satre M, Chen L, Soprano K and
Kochar D (1994) A sustained elevation in retinoic acid receptor-beta2 mRNA and
protein occures during retinoic-acid induced fetal dysmorphogenesis. Mech of
Dev 45: 243 ± 253
Sulik K, Sulik KK, Cook C and Webster W (1988) Teratogenes and craniofcial
malformations Develpoment 102: suppl, 213 ± 231
Sulik KK and Dehart DB (1988) Retinoic-acid-induced limb malformations resulting
from apical ectodermal ridge cell death. Teratology 37: 527 ± 537
Szondy Z, Reichert U, Branardon J, Michel S, Toth R, Ancian P, Ajzner E and Fesus L
(1997) Induction of apoptosis by retinoids and RARg selective compounds in
mouse thymocytes through a novel apoptosis pathway. Mol Pharmacol 51: 972 ±
982
Taneja R, Bouillet P, Boylan J, Baub M, Roy L, Gudas L and Chambon P (1995)
Reexpression of retinoic acid receptor gamma or overexpression of RAR-alpha
or RAR-beta in RAR-gamma null F9 cells reveals a partial functional redundancy
between the three RAR types. Proc Natl Acad Sci USA 92: 7854 ± 7858
Thomazy V and Fesus L (1989) Differential expression of tissue transglutaminase in
human cells. An immunohistochemical study. Cell Tissue Res 255: 215 ± 224
Verma AK, Shoemaker A, Simsiman R, Denning M and Zachman RD (1992)
Expression of retinoic acid nuclear receptors and tissue transglutaminase is
altered in various tissues of rats fed a vitamin A-deficient diet. J Nutr 122: 2144 ±
2152
Watson R, Rothstein O, Parodo J, Bitar R, Hackam D and Marshall J (1997)
Granulocytic differentiation of HL-60 cells results in spontaneous apoptosis
mediated by increased caspase expression. FEBS Lett. 412: 603 ± 609
Wilson J, Roth C and Warkany J (1953) An analysis of the syndrome of malformations
induced by maternal vitamin A deficiency. Effects of restoration of vitamin A at
various times during gestation. Am J of Anatomy 92: 189 ± 217
Wolbach S and Howe P (1925) Tissue changed following deprivation of fat-soluble
vitamin A. Journal of Exp Med 42: 753 ± 780
Yang Y, Vaccio M and Ashwell J (1993) 9-cis retinoic acid inhibits activation-driven Tcell apoptosis: implications for retinoid x receptor involvement in thymocyte
development. Proc Natl Acad Sci USA 90: 6170 ± 6174
Zakeri ZF and Ahuja HS (1994) Apoptotic Cell Death In the Limb and Its Relationship
to Pattern Formation. Biochemistry & Cell Biology 72: 603 ± 613
Zhang LX, Mills KJ, Dawson MI, Collins S J and Jetten AM (1995) Evidence for the
involvement of retinoic acid receptor RAR alpha-dependent signaling pathway in
the induction of tissue transglutaminase and apoptosis by retinoids. J Biol Chem
270: 6022 ± 6029